Download A046010107

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Expectation–maximization algorithm wikipedia , lookup

Nonlinear dimensionality reduction wikipedia , lookup

K-nearest neighbors algorithm wikipedia , lookup

Nearest-neighbor chain algorithm wikipedia , lookup

Cluster analysis wikipedia , lookup

K-means clustering wikipedia , lookup

Transcript
Ch. Mallikarjuna Rao et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 6( Version 1), June 2014, pp.01-07
RESEARCH ARTICLE
www.ijera.com
OPEN ACCESS
Crop Yield Analysis of the Irrigated Areas of All Spatial
Locations in Guntur District of AP.
Ch. Mallikarjuna Rao1, Dr A. Ananda Rao2
(Department of computer science and Engineering, Asso. Prof GRIET)
(Department of Computer Science and Engineering, Professor, JNTUA)
ABSTRACT
Spatial data mining is a process to discover interesting and potentially useful spatial patterns embedded in
spatial databases, which are voluminous in their sizes. Efficient techniques for extracting information from geospatial data sets can be of importance to various sectors like research, defense and private organizations for
generating and managing large geo-spatial data sets. The current approach towards solving spatial data mining
problems is to use classical data mining techniques. Effective analysis was done using the hybrid data mining
techniques by mixing both clustering and classification techniques. In this paper crop yield of spatial locations
of Guntur district were taken and studied using the hybrid technique.
Keywords: geo-spatial data sets, hybrid data mining technique, clustering, classification, spatial locations
I.
Introduction
Indian agriculture is known for more
fluctuations in terms of crop yield, crop output and
crop intensity. Despite growth in Technology and
irrigation the production and income are highly
instable [11]. Guntur district is located in the state of
Andhra Pradesh. It is situated along the east coast of
the Bay of Bengal. Its coastline is approximately 100
kilometers. It is the largest city and is the
administrative center of the District. It has 57
mandals starting from Guntur, Pedakakani,
Prathipadu, etc..to ending with Nizampatnam. Major
crops are Cotton and Chillies. Analysis need to be
done on the agriculture data sets which requires
classical data mining techniques apart from statistical
techniques. The data mining techniques[12] like
classification, clustering and association are required
to apply on the realistic data sets for analysis and
conclusions on the agriculture crop yields [8, 9, and
10] of various seasons like Kharif and rubby. The
following existing techniques are discussed along
with proposed hybrid approach.
II.
Literature Survey
K-Means clustering Algorithm:The k-means
algorithm [3] [Hartigan& Wong 1979] is the well
known clustering technique used in scientific and
industrial applications. Its name comes from centroid
which is the mean of c of k clusters C. This technique
is not suitable for categorical attributes. It is more
suitable for numerical attributes. K-means[1]
algorithm uses squared error criteria and is
extensively used algorithm. The data is partitioned
into K clusters (C1;C2; : : : ;CK), using this algorithm
which are represented by their centers or means. The
www.ijera.com
mean of all the instances belonging to that cluster
gives the center of each cluster.
The pseudo-code of the K-means algorithm
is given by Fig.1. The algorithm begins with
randomly selected initial set of cluster centers. Each
instance is assigned to its closest cluster center for
every iteration based on Euclidean distance
calculated between the two. After thatre-calculate the
cluster centers.
the instances number belonging to cluster k is
denoted byNkand the mean of the cluster kis
represented by µk.
There is a possibility for a number of
convergence conditions. If the partitioning error is
not reduced by the relocation of the centers the search
will be stopped as it concludes that the present
partition is locally optimal. Another stopping
criteriais that if it exceeds a pre-defined number of
iterations.
Input: S (instance set), K (no. of clusters)
Output: clusters
1: Initialize K cluster centers.
2: while termination condition is not satisfied do
3: Assign instances to the nearest cluster center.
4: Update cluster centers based on the assignment.
5: end while
Figure1. K-means Algorithm
This algorithm may be viewed as a gradientdecent procedure. In which it begins with random
selection of an initial set number of K cluster-centers
and iteratively updates by which error function will
decrease. A rigorous proof of the finite convergence
1|P age
Ch. Mallikarjuna Rao et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 6( Version 1), June 2014, pp.01-07
of the K-means type algorithms is given in [4]. The
complexity of T iterations of the K-means algorithm
performed on a sample size of m instances, each
characterized by N attributes, is: O(T * K * m * N).
This linear complexity is one of the reasons for the
popularity of the K- means algorithms. Even if the
number of instances is substantially large (which
often is the case nowadays), this algorithm is
computationally attractive. Thus, the K-means
algorithm has an advantage in comparison to other
clustering methods (e.g. hierarchical clustering
methods), which have non-linear complexity. Other
reasons for the algorithm’s popularity are its ease of
interpretation, simplicity of implementation, speed of
convergence and adaptability to sparse data [5]. The
Achilles heel of the K-means algorithm involves the
selection of the initial partition. The algorithm is very
sensitive to this selection, which may make the
difference between global and local minimum. Being
a typical partitioning algorithm, the K-means
algorithm works well only on data sets having
isotropic clusters, and is not as versatile as single link
algorithms, for instance.
In addition, this algorithm is sensitive to
noisy data and outliers (a single outlier can increase
the squared error dramatically); it is applicable only
when mean is defined (namely, for numeric
attributes); and it requires the number of clusters in
advance, which is not trivial when no prior
knowledge is available. The use of the K-means
algorithm is often limited to numeric attributes.
The similarity measure for numeric attributes and
categorical attributes differ in the first case take the
square Euclidean distance; where as in second case
take the number of mismatches between objects and
the cluster prototypes.
Another partitioning algorithm, which
attempts to minimize the SSE is the K-medoids or
PAM (partition around medoids[2]). It is similar to
the K-means algorithm. It differs with k-means
mainly in representation of the different clusters.
Each cluster is represented by the most centric object
in the cluster, rather than by the implicit mean that
may not belong to the cluster. This method is more
robust than K-means algorithm even if there is noise
and outliers. The influence of outliers is less. The
processing of K-means is less costlierwhen compared
to this. It is required to declare the number of clusters
K for both the methods.It is not mandatory to use
only SSE. Estivill-Castro (2000) analyzed the total
absolute error criterion. He stated that it is better to
summing up the absolute error instead of summing
up the squared error. This method is superior in terms
of robustness, but it requires more computational
effort. The objective function is defined as the sum of
discrepancies between a point and its centroid which
is expressed through an appropriate distance. The
total intra-cluster variance is the norm based
www.ijera.com
www.ijera.com
objective function, and is defined as the sum of
squares of errors between the points and the
corresponding centroids,
2
E( C ) = 𝑘𝑗=1 𝑥 𝑖 ∈𝐶𝑗 𝑥𝑖 − 𝑐𝑗
It can be rationalized as log-likelihood for
normally distributed mixture model. In Statistic it is
extensively used. Therefore, it is derived from
general framework of probability. Its iterative
optimization has two versions. The first technique is
similar to EM algorithm one of the Data mining
technique. It has two-stepsof major iterations in
which first time reassigns all the points to their
nearest centroids, and second timerecompute
centroids of newly formed groups. Iterations continue
until a stopping criterion is achieved (for example, no
reassignments happen). This version is known as
Forgy.s algorithm [6] and has many advantages:

It easily works with any -norm p L

It allows straightforward parallelization[5]

It is insensitive with respect to data ordering.
If a move has a positive effect, the point is
relocated and the two centroids are recomputed.
J48 algorithm:J48 [7]is the optimization of C4.5
algorithm and also animprovised versions of it.Its
output result is a Decision tree which is of a tree
structure. It has a root node, along with intermediate
nodes apart from leaf nodes. Except Root node and
leaf nodes every other node in the tree consistsof
decision and which leads to our result. This tree
divides the given space of a data set into areas of
mutually exclusive. In this its data points are
described by every area will have a label, a value or
an action.The criterion of splitting is used to find
which attribute makes the best split on the portion of
tree of the training data set which in turn reaches to a
particular node. Decision tree is formed by using the
children attribute of the data set.
III. Proposed Approach
The raw data set was converted to the required format
and then apply the data mining technique namely kmeans clustering algorithm and then we get the new
data set namely clustered data set. The classification
technique J48 was applied on that clustered data set
which results in hybrid model.
IV. Implementation of proposed
approach
A data set on Irrigation for the Year 200708& 2011-12 of Guntur district were considered for
analysis with k-means (or simple k-means) clustering
data [8,9] mining technique having 15 attributes
namely Mandal, latitude, latitude1, longitude,
longitude1, crop, area_irrigated, area_unirrigated, etc.
Clustering is an Unsupervised one and here Missing
values are replaced with mean and mode apart from
2|P age
Ch. Mallikarjuna Rao et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 6( Version 1), June 2014, pp.01-07
Discretization with 10 bins. Here full training set as
the clustering model. The cluster table was
represented by Table 1 and Cluster proximities were
given by Table2. The cluster graph was shown below
in Fig.2. The 5 clusters are shown in 4 different
colors with instances on x-axis and mandals on yaxis. A hybrid model was developed and applied after
applying K-means clustering.We took the J48
classification technique on the obtained clustered
data set. The result was summarized as follows. It is
with Minimum confidence is 0.25 and binary tree
classifier. It also has same number of instances
i.e.114 with 19 attributes. The attributes are
Instance_number, mandal, latitude, latitude1,
longitude,
longitude1,
crop,
area_irrigated,
area_unirrigated, area_total, etc. attributes. The J48
pruned tree is given below.
crop = Cotton
| latitude<= 16.37184
| | area_unirrigated<= 3947: cluster4 (34.0/2.0)
| | area_unirrigated> 3947: cluster3 (6.0)
| latitude> 16.37184: cluster3 (17.0/1.0)
crop = Chillies
| pr
oductivity_unirrigated<= 3.239
| | latitude <= 16.340198
| | | production_irrigated<= 11367.5: cluster2
(35.0/1.0)
| | | production_irrigated> 11367.5: cluster1
(2.0)
| | latitude > 16.340198: cluster1 (15.0)
| productivity_unirrigated> 3.239: cluster0 (5.0)
The pruned tree in tree structure was shown in Fig.3
with crop as root node and clusters as leaf nodes.
Correctly Classified Instances are 103 (90.351%) and
Incorrectly Classified Instances are 11 (9.649%).
Kappa statistic value is 0.8716.Kappa is a chancecorrected measure of agreement between the
classifications and the true classes. It's calculated by
taking the agreement expected by chance away from
the observed agreement and dividing by the
maximum possible agreement. A value greater than 0
means that your classifier is doing better than chance.
The Kappa statistic is near to 1 so it is near to perfect
agreement. Confusion Matrix: In the confusion
matrix each column represents instances in the
predicted class and each row indicates instances in
actual class. These mining techniques were applied
on various data sets.
www.ijera.com
16.4542 N, Longitude is 80.0421 E with crop Chillies
and area irrigated is 1531 area un-irrigated is 710
with area total is 1946 with highest productivity
_irrigated 5.5 and productivity_unirrigated is 5. It
also has highest production_irrigated and lowest
Production_unirrigated with Productivity _total and
Production_total as 0.
Cluster1 and 2 are also chillies as crop and
production-irrigated is highest in cluster1with
mandalsMangalagiri and chebrolu. Cluster 3 and 4
are with Mandals Guntur and nagaram and cotton is
their crop. Cluster 3 is with highest area-irrigated and
area_unirrigated. It also has highest area_total and
production-unirrigated among the clusters. Confusion
matrix shows that cluster 0 was classified accurately
next to that was cluster 2. In cluster 1 out of 18
instances 23 were classified under cluster 2 and 1 was
classified under cluster 3. Cluster3 and4 are not
classified correctly. The values of TP rate and Fmeasure for cluster 0 are 1 indicates they are
classified accurately. Cluster2 and 4 are also has TP
rate and F-measure are almost 1,so classified almost
accurately.
Fig.4 and Fig.5 are showing that the
irrigated areas of major crops cotton and chillis
across mandals are increasing from the years 2007-08
to the years 2011-12. In these graphs number of
mandals were taken along x-axis and Irrigated area
along y-axis. Cotton area in Hectors is higher than
Chillis area in Hectors shown in Fig.6. in this graph
name of the mandalswere taken along x-axis and
Area in Hectors is taken along y-axis. Interesting
measures were found when Association rule Apriori
was applied on the clustered spatial data set and the
rules were shown in fig.7 and fig.8. Initially
discretization was done on the clustered data set as a
hybrid technique later apriori association rule was
applied. It was found that as we increase min support
from 0.1to 0.4 and to 0.6 and then to 0.8 keeping
minimum support to 0.9 constant the Latitude 1 = N,
Longitude 1 = E, and Productivity_total are
associated with 114 instances and along with the
other attribute productivity_irrigated are together
formed association rules with 99 instances. When
minimum support and minimum confidence attains
values both 0.9 or either of them takes the values 0.9
or 1.0 vice versa or both takes values as 1.0 the
attributes longitude1(E), latitude 1(N) and
Productivity_total are only associated together.
V. Results & Analysis:
Initially k-means clustering technique was
applied on Guntur Irrigation data set for analysis.
Cluster 0: It was found that for the
PedanandipaduMandal centroids of the latitude is
www.ijera.com
3|P age
Ch. Mallikarjuna Rao et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 6( Version 1), June 2014, pp.01-07
www.ijera.com
Cluster centroids:
Cluster#
Attribute
Full Data
0
1
2
3
4
(114)
(5)
(18)
(35)
(23)
(33)
mandal
Guntur Pedanandipadu Mangalagiri
Chebrolu Guntur
Nagaram
latitude
16.2535 16.4542
16.4478
16.1354 16.4331
16.1173
latitude1
N
N
N
N
N
N
longitude
80.2683
80.0421
80.0966
80.3918 80.0998
80.3826
longitude1
E
E
E
E
E
E
crop
Cotton
ChilliesChilliesChillies Cotton
Cotton
area_irrigated
1476.8222 1531.7644 1737.3037
1108.6413 1901.1014 1421.202
area_unirrigated
1995.4521 710.4
970.7976
1391.1284 4536.2805
1619.1283
area_total
2566.5663 1946.8
2179.3889
1485.2265 5499.0681 1974.6685
productivity_irrigated 5.186
5.531
5.4436
5.4042 4.7033
5.0984
productivity_unirrigated 2.7223
5.0448
2.7092
2.7223
2.2873
2.6809
productivity_total
0
0
0
0
0
0
production_irrigated
11367.5 11772.9
13594.4167
10537.5143 11100.7391 11157.6061
production_unirrigated 9012.8438 3560
8729.3524
9012.8438 11117.8492 8526.5388
production_total 0
0
0
0
0
0
Table 1: Cluster table on Irrigation data set
Fig.2: Cluster graph
Fig.3: Pruned tree
www.ijera.com
4|P age
Ch. Mallikarjuna Rao et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 6( Version 1), June 2014, pp.01-07
Cluster-I
5 1.682367
4 1.682367
2 1.682367
3 1.682367
Cluster-II
1 1.682367
3 2.236069
4 2.236069
5 2.236069
Cluster-III
1 1.682367
2 2.236069
4 2.236069
5 2.236069
Cluster-IV
1 1.682367
3 2.236069
2 2.236069
5 2.236069
www.ijera.com
Cluster - V
1 1.682367
3 2.236069
4 2.236069
2 2.236069
Table2. Clusters along with their Proximities
The Confusion Matrix is given as below
a bc d e <-- classified as
5 0 0 0 0 | a = cluster0
0 14 3 1 0 | b = cluster1
0 0 34 0 1 | c = cluster2
0 2 0 20 1 | d = cluster3
0 0 1 2 30 | e = cluster4
6000
Cotton Irrigated area across years
Irrigated Area
5000
Irrigated 2007-08
Irrigated 2011-12
4000
3000
2000
1000
0
-1000
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57
Numbeer of Mandals
Fig4: Cotton irrigated area across years and mandals
8000
Chillies Irrigated area across years
Irrigated area
6000
Irrigated 2007-08
Irrigated 2011-12
4000
2000
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57
Number of Mandals
Fig.5: Chillis irrigated area across years and mandals
www.ijera.com
5|P age
Ch. Mallikarjuna Rao et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 6( Version 1), June 2014, pp.01-07
12000
Cotton & Chilli area in Hec. across mandals
10000
www.ijera.com
Total cotton area
Total Chillies area
Area in Hectors
8000
6000
4000
0
Guntur
Pedakakani
Prathipadu
Vatticheruk…
Pedanandip…
Mangalagiri
Tadepalli
Thulluru
Tadikonda
Amaravathi
Sattenapalli
Pedakurapa…
Medikonduru
Phirangipur…
Muppalla
Krosuru
Atchampet
Rajupalem
Bellamkonda
Narsaraopeta
Rompicherla
Chilakalurip…
Nadendla
Edlapadu
Vinukonda
Nuzendla
Ipuru
Bollapalli
Savalyapuram
Piduguralla
Machavaram
Dachepalli
Karampudi
Nekarikallu
Macherla
Veldurthi
Durgi
Rentachintala
Gurazala
Tenali
Duggirala
Vemuru
Kollipara
Kolluru
Ponnur
Tsunduru
Chebrolu
Amartaluru
Bapatla
Karlapalem
P.V.Palem
Kakumanu
Repallle
Nagaram
Bhattiprole
Cherukupalli
Nizampatnam
2000
Names of the Mandals
Fig.6: Cotton and Chilli area in Hectors across Mandals of Guntur
Best rules found:
1. longitude1=E 114 ==> latitude1=N 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
2. latitude1=N 114 ==> longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
3. productivity_total='All' 114 ==> latitude1=N 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
4. latitude1=N 114 ==>productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
5. productivity_total='All' 114 ==> longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
6. longitude1=E 114 ==>productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
7. longitude1=E productivity_total='All' 114 ==> latitude1=N 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
8. latitude1=N productivity_total='All' 114 ==> longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
9. latitude1=N longitude1=E 114 ==>productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
10. productivity_total='All' 114 ==> latitude1=N longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
11. longitude1=E 114 ==> latitude1=N productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
12. latitude1=N 114 ==> longitude1=E productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
13. productivity_irrigated='(4.967-5.5112]' 99 ==> latitude1=N 99 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
14. productivity_irrigated='(4.967-5.5112]' 99 ==> longitude1=E 99 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
15. productivity_irrigated='(4.967-5.5112]' 99 ==>productivity_total='All' 99 <conf:(1)> lift:(1) lev:(0) [0]
conv:(0)
16. longitude1=E productivity_irrigated='(4.967-5.5112]' 99 ==> latitude1=N 99 <conf:(1)> lift:(1) lev:(0)
[0] conv:(0)
17. latitude1=N productivity_irrigated='(4.967-5.5112]' 99 ==> longitude1=E 99 <conf:(1)> lift:(1) lev:(0)
[0] conv:(0)
18. productivity_irrigated='(4.967-5.5112]' 99 ==> latitude1=N longitude1=E 99 <conf:(1)> lift:(1) lev:(0)
[0] conv:(0)
19. productivity_irrigated='(4.967-5.5112]' productivity_total='All' 99 ==> latitude1=N 99 <conf:(1)> lift:(1)
lev:(0) [0] conv:(0)
20. latitude1=N productivity_irrigated='(4.967-5.5112]' 99 ==>productivity_total='All' 99 <conf:(1)> lift:(1)
lev:(0) [0] conv:(0)
Fig.7 Association rules when minimum support = 0.85 and minimum confidence =0.9
Best rules found:
1. longitude1=E 114 ==> latitude1=N 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
2. latitude1=N 114 ==> longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
3. productivity_total='All' 114 ==> latitude1=N 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
4. latitude1=N 114 ==>productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
5. productivity_total='All' 114 ==> longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
6. longitude1=E 114 ==>productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
7. longitude1=E productivity_total='All' 114 ==> latitude1=N 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
8. latitude1=N productivity_total='All' 114 ==> longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
9. latitude1=N longitude1=E 114 ==>productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
10. productivity_total='All' 114 ==> latitude1=N longitude1=E 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
11. longitude1=E 114 ==> latitude1=N productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
12. latitude1=N 114 ==> longitude1=E productivity_total='All' 114 <conf:(1)> lift:(1) lev:(0) [0] conv:(0)
Fig.7 Association rules when minimum support = 1.0 and minimum confidence =1.0
www.ijera.com
6|P age
Ch. Mallikarjuna Rao et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 6( Version 1), June 2014, pp.01-07
VI. Conclusion
There is a correlation between the cotton
irrigated area of Irrigation 2007-08 and
Irrigation2011-12 data and its value is 0.753871.
There is a higher correlation between chillies
irrigated area of Irrigation 2007-08 and
Irrigation2011-12
data
and
its
value
is
0.869333.These analysis are specified in the results
and analysis. Future scope of this hybrid approach
can be extended to various agricultural spatial
locations and also to various agricultural yields for
effective analysis .
[11]
www.ijera.com
No: 459-465, ISBN No: 978-93-35107-1495
Ramesh Chand, S.S. Raju, Instability in
Andhra Pradesh agriculture -A Disaggregate
Analysis, Agricultural Economics Research
Review Vol. 21 July-December 2008 pp283288.
[12] D. Hand, el al., Principles of Data
Mining. Massachusetts: MIT Press, 2001.
AUTHORS
References
[1]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
PavelBerkhin, Survey of Clustering Data
Mining Techniques, Accrue Software, Inc.
[2] Kaufman, L. and Rousseeuw, P.J., 1987,
Clustering by Means of Medoids, In Y.
Dodge, editor, Statistical Data Analysis,
based on the L1 Norm, pp. 405- 416,
Elsevier/North Holland, Amsterdam.
Hartigan, J. A. Clustering algorithms. John
Wiley and Sons., 1975.
Selim, S.Z., and Ismail, M.A. K-means-type
algorithms: a generalized convergence
theorem and characterization of local
optimality. In IEEE transactions on pattern
analysis and machine learning, vol. PAMI-6,
no. 1, January, 1984.
Dhillon I. and Modha D., Concept
Decomposition for Large Sparse Text Data
Using Clustering.Machine Learning.42,
pp.143-175. (2001).
FORGY, E. 1965. Cluster analysis of
multivariate
data:
Efficiency versus
interpretability of classification. Biometrics,
21, 768-780.
Yugalkumar and G. Sahoo, Analysis of
Bayes, Neural Network and Tree Classifier
of Classification Technique in Data Mining
using WEKA.
Dr.T. V. RajiniKanth, Ananthoju Vijay
Kumar, Estimation of the Influence of
Fertilizer Nutrients Consumption on the
Wheat Crop yield in India- a Data mining
Approach, 30 Dec 2013, Volume 3, Issue 2,
Pg.No: 316-320, ISSN: 2249-8958 (Online).
Dr.T. V. RajiniKanth, Ananthoju Vijay
Kumar, A Data Mining Approach for the
Estimation of Climate Change on the Jowar
Crop Yield in India, 25Dec2013,Volume 2
Issue 2, Pg.No:16-20, ISSN: 2319-6378
(Online).
A. Vijay Kumar,
“Estimation of the
Influential Factors of rice yield in India” 2 nd
International Conference on Advanced
Computing methodologies ICACM-2013,
02-03 Aug 2013, ElsevierPublications, Pg.
www.ijera.com
Ch.MallikarjunaRao
Received his B.Tech degree in computer
Science and engineering from Dr.Baba sahib
AmbedkarMarathwada University, Aurangabad,
Maharastra in 1998,and M.Tech Degree in Computer
Science and Engineering from J.N.T.U Anantapur
,Andhrapradesh in 2007. He is currently pursuing his
Ph.D degree from JNTU Ananthapur University,
Andhra Pradesh. Currently he is working as
Associate Professor in the department of Computer
Science and Engineering of GokarajuRangaraju
Institute of Engineering and Technology, Hyderabad,
India.His research interest includes Data bases and
data mining.
Dr. AnandaRaoAkepogureceived his B.
TechdegreeinComputerScience&Engineerin
g from University of Hyderabad, Andhra
Pradesh, India and M.Tech degree in A.I &
Robotics from University of Hyderabad,
Andhra Pradesh, India. He received Ph.D degree from
Indian Institute of Technology Madras, Chennai, India.
He is Professor of ComputerScience& Engineering
Department and currently working as Principal of
JNTUA College of Engineering, Anantapur, Jawaharlal
Nehru Technological University, Andhra Pradesh, India.
Dr. Rao published more than 100 publications in various
National and International Journals/ Conferences. He
received Best Research Paper award for the paper
titled “An Approach to Test Case Design for Cost
Effective Software Testing” in an International
Conference on Software Engineering held at Hong
Kong, 18-20 March 2009. He also received Best
Educationist Award for outstanding achievements in
the field of education by International Institute of
Education & Management, New Delhi on 21st Jan.
2012. He bagged Bharat VidyaShiromani Award
from
Indian
Solidarity
Council
and
RashtriyaVidyaGaurav Gold Medal Award from
International Institute of Education & Management,
New Delhi on 19th March, 2012. Dr.Rao got Best
Computer Science and Engineering Faculty award from
ISTE for the Year 2013. His main research interest
includes software engineering and data mining.
7|P age